CT-Derived Endocranial Volume as a Surrogate for MRI-Derived Total Brain Volume in Domestic Cats

<title>Abstract</title> <p> <bold>Background</bold> Magnetic resonance imaging (MRI) is the gold standard for brain volumetry but is often constrained by cost and limited accessibility in veterinary medicine. Computed tomography (CT) offers a faster and more available alternative. To develop a reproducible 3D Slicer workflow for feline endocranial volume (EV) and to quantify the agreement between CT-derived EV and MRI-derived total brain volume (TBV) to justify CT as a clinical surrogate, head CT and MRI scans of fifteen domestic cats (Felis catus) of various breeds, aged 4 months to 13 years, were evaluated in this retrospective observational study. Brain volumes were segmented using 3D Slicer, where MRI-TBV was measured via manual ROI annotation, and CT-EV was calculated using thresholding and seeded region-growing. <bold>Results</bold> CT-derived EV strongly predicted MRI-derived total brain volume (R2 = 0.925, p &lt; 0.00002). EV was the only significant predictor (β = 1.04, p &lt; 0.001). LOOCV confirmed robust accuracy (RMSE: 1.11 cm ³ ; mean absolute error: 1.04 cm ³ ), representing approximately 3%–4% of mean TBV. Bland-Altman analysis showed negligible mean bias (0.00 cm ³ ) with narrow 95% limits of agreement (95% CI: − 1.54 to + 1.54 cm ³ ). <bold>Conclusions</bold> CT-derived endocranial volume is a highly accurate surrogate for MRI-derived brain volume in domestic cats. This validated workflow enables cost-effective volumetric assessments in clinical neurology and comparative research where MRI is unavailable.Keywords: Brain volume, Computed tomography, Magnetic resonance imaging, Feline neuroanatomy, 3D Slicer, Endocranial volume (EV). </p>